Search NASA⌕ Search

SEARCH · Search NASA

Results for “deep data”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 631 records · Page 35

Field Data Analysis and Modeling of Drillstring Vibrations to Identify Inefficiency in Deep Geothermal Drilling

One challenge that hinders efficient drilling and causes downhole tool failures is severe drillstring vibrations. The objective of this paper is to identify the root cause of drillstring vibrations in deep geothermal wells and investigate their effect on drilling performance using data analytics and vibration modeling. A near-bit sub was utilized to collect vibration data, where the burst data was used to obtain the drillstring torsional natural frequency. The data showed that the highest lateral acceleration and stick-slip severities (SSS) occur at the higher mechanical specific energy (MSE) range, while the low vibration levels were encountered in the optimum range of the MSE curve. The elevated SSS levels consistently occur at the low to mid-range of applied RPM and the highest sonic velocity. Additionally, the lateral acceleration levels are higher at the low to mid-range of the applied RPM. The effect of drillstring vibration on drilling performance was recognized by the presented MSE analysis.

15 GEOTHERMAL ENERGY↗

Data supporting manuscript from L. Sheneman, G. Stephanopoulos, A.E. Vasdekis titled "Deep learning classification of lipid droplets in quantitative phase images" as currently under review at PLOS ONE. This includes: 1) raw and binary labeled Quantitative Phase Images (QPI) of Y. lipolytica cells used in the analyses described within the manuscript. 2) various derived data including classifier scores, etc.

Data supporting manuscript from L. Sheneman, G. Stephanopoulos, A.E. Vasdekis titled "Deep learning classification of lipid droplets in quantitative phase images" as currently under review at PLOS ONE. This includes: 1) raw and binary labeled Quantitative Phase Images (QPI) of Y. lipolytica cells used in the analyses described within the manuscript. 2) various derived data including classifier scores, etc.

ANN↗

Reynolds-Averaged Turbulence Modeling Using Deep Learning with Local Flow Features: An Empirical Approach

Reynolds-Averaged Navier-Stoke (RANS) models offer an alternative avenue in predicting flow characteristics when the corresponding experiments are difficult to achieve due to geometry complexity, limited budget, or knowledge. RANS models require the knowledge of subgrid scale physics to solve conservation equations for mass, energy, and momentum. Mechanistic turbulence models, such as k-epsilon, are generally evaluated and calibrated for specific flow conditions with various degrees of uncertainty. These models have limited capability to assimilate a substantial amount of data due to model form constraints. Meanwhile, deep learning (DL) has been proven to be universal approximators with the potential to assimilate available, relevant, and adequately evaluated data. Moreover, deep neural networks (DNNs) can create surrogate models without knowing function forms. Such a data-driven approach can be used in updating fluid models based on observations as opposed to hard-wiring models with precalibrated correlations. The paper presents progress in applying DNNs to model Reynolds stress using two machine learning (ML) frameworks. A novel flow feature coverage mapping is proposed to quantify the physics coverage of DL-based closures. It can be used to examine the sufficiency of training data and input flow features for data-driven turbulence models. The case of a backward-facing step is formulated to demonstrate that not only can DNNs discover underlying correlation behind fluid data but also they can be implemented in RANS to predict flow characteristics without numerical stability issues. Finally, the presented research is a crucial stepping-stone toward the data-driven turbulence modeling, which potentially benefits the design of data-driven experiments that can be used to validate fluid models with ML-based fluid closures.

42 ENGINEERING↗

DEEP CELLULAR RECURRENT NEURAL ARCHITECTURE FOR EFFICIENT MULTIDIMENSIONAL TIME-SERIES DATA PROCESSING

Efficient processing of time series data is a fundamental yet challenging problem in pattern recognition. Though recent developments in machine learning and deep learning have enabled remarkable improvements in processing large scale datasets in many application domains, most are designed and regulated to handle inputs that are static in time. Many real-world data, such as in biomedical, surveillance and security, financial, manufacturing and engineering applications, are rarely static in time, and demand models able to recognize patterns in both space and time. Current machine learning (ML) and deep learning (DL) models adapted for time series processing tend to grow in complexity and size to accommodate the additional dimensionality of time. Specifically, the biologically inspired learning based models known as artificial neural networks that have shown extraordinary success in pattern recognition, tend to grow prohibitively large and cumbersome in the presence of large scale multi-dimensional time series biomedical data such as EEG. Consequently, this work aims to develop representative ML and DL models for robust and efficient large scale time series processing. First, we design a novel ML pipeline with efficient feature engineering to process a large scale multi-channel scalp EEG dataset for automated detection of epileptic seizures. With the use of a sophisticated yet computationally efficient time-frequency analysis technique known as harmonic wavelet packet transform and an efficient self-similarity computation based on fractal dimension, we achieve state-of-the-art performance for automated seizure detection in EEG data. Subsequently, we investigate the development of a novel efficient deep recurrent learning model for large scale time series processing. For this, we first study the functionality and training of a biologically inspired neural network architecture known as cellular simultaneous recurrent neural network (CSRN). We obtain a generalization of this network for multiple topological image processing tasks and investigate the learning efficacy of the complex cellular architecture using several state-of-the?art training methods. Finally, we develop a novel deep cellular recurrent neural network (CDRNN) architecture based on the biologically inspired distributed processing used in CSRN for processing time series data. The proposed DCRNN leverages the cellular recurrent architecture to promote extensive weight sharing and efficient, individualized, synchronous processing of multi-source time series data. Experiments on a large scale multi-channel scalp EEG, and a machine fault detection dataset show that the proposed DCRNN offers state-of-the-art recognition performance while using substantially fewer trainable recurrent units.

Vidyaratne, Lasitha S.↗

Power System Event Classification and Localization Using a Convolutional Neural Network

Detection and timely identification of power system disturbances are essential for situation awareness and reliable electricity grid operation. Because records of actual events in the system are limited, ensemble simulation-based events are needed to provide adequate data for building event-detection models through deep learning; e.g., a convolutional neural network (CNN). An ensemble numerical simulation-based training data set have been generated through dynamic simulations performed on the Polish system with various types of faults in different locations. Such data augmentation is proven to be able to provide adequate data for deep learning. The synchronous generators’ frequency signals are used and encoded into images for developing and evaluating CNN models for classification of fault types and locations. With a time-domain stacked image set as the benchmark, two different time-series encoding approaches, i.e., wavelet decomposition-based frequency-domain stacking and polar coordinate system-based Gramian Angular Field (GAF) stacking, are also adopted to evaluate and compare the CNN model performance and applicability. The various encoding approaches are suitable for different fault types and spatial zonation. With optimized settings of the developed CNN models, the classification and localization accuracies can go beyond 84 and 91%, respectively.

24 POWER TRANSMISSION AND DISTRIBUTION↗

A Coupled Deep Learning Model for Estimating Surface NO 2 Levels from Remote Sensing Data: 15-Year Study Over the Contiguous United States

This study proposes a novel two-step deep learning (DL) model for estimating surface NO 2 concentrations using satellite data over the contiguous United States (CONUS) from 2005 to 2019. The first phase of the model uses partial convolutional neural network (PCNN), an advanced DL model that accurately imputes gaps between surface NO 2 stations and creates 5,478 daily-mean NO 2 grids (PCNN-NO 2 ) of the 2005-2019 period over the study area. We then feed the PCNN-NO 2 , along with other predictor variables, into a deep neural network (DNN) to estimate surface NO 2 levels, achieving exceptional performance with a correlation coefficient of 0.975 to 0.978, a mean absolute bias of 0.99 ppb to 1.38 ppb, and a root mean square error of 1.47 ppb to 1.97 ppb. Spatial cross-validation results also indicate strong spatial performance of PCNN-DNN surface NO 2 estimates. In addition to its accurate estimates, the PCNN-DNN model consistently generates estimated NO 2 grids without any missing values, improving the quality of various applications such as emission reduction strategies and public health studies. Between 2005 and 2019, the 5,478 daily estimated NO 2 grids over the CONUS reveal significant reductions in NO 2 levels in fourteen major urban environments: Washington D.C. (-43%), New York (-45%), Los Angeles (-38%), Chicago (-25%), Boston (-43%), Houston (-34%), Dallas (-40%), Philadelphia (-41%), Phoenix (-38%), Detroit (-20%), Denver (-23%), Atlanta (-0.7%), Cincinnati (-38%), and Pittsburgh (-56%). Furthermore, the study shows that the denser urban regions that in-situ stations are installed in, the higher the difference between in-situ observations and regional-mean NO 2 levels.

54 ENVIRONMENTAL SCIENCES↗

A progressive deep transfer learning approach to cycle-skipping mitigation in FWI, In SEG Technical Program Expanded Abstracts 2019, pp. 2348-2352. Society of Exploration Geophysicists, 2019.

To effectively mitigate the cycle-skipping phenomenon in full waveform inversion (FWI), we developed an inception-based deep learning neural network to reconstruct the absent low frequency data by exploiting the subsurface low wavenumber information buried in the acquired high frequency data. Two unique features of our deep learning network for the low frequency data reconstruction are 1) Dual Data Feed; 2) Progressive Transfer Learning. With the Dual Data Feed feature, not only the high frequency data, but also the corresponding Beat Tone data are fed into the neural network, significantly reducing the network complexity and the network training cost. The Progressive Transfer Learning technique enables us to initiate the deep learning network with only one training dataset. Unlike other deep learning approaches in this category, our training dataset is not fixed. Instead, the training dataset and the network accuracy are progressively enhanced in an iterative manner within the Progressive Transfer Learning workflow. The numerical experiments validated that the low frequency data reconstructed by this progressive transfer learning network are sufficiently accurate for a successful FWI inversion without suffering from the cycle-skipping without any priori geological information.

58 GEOSCIENCES↗

Millimeter-wave observations of Euclid Deep Field South using the South Pole Telescope: A data release of temperature maps and catalogs

Context. The South Pole Telescope third-generation camera (SPT-3G) has observed over 10,000 square degrees of sky at 95, 150, and 220 GHz (3.3, 2.0, 1.4 mm, respectively) and will significantly overlap the ongoing 14,000 square-degree Euclid Wide Survey. The Euclid collaboration recently released Euclid Deep Field South (EDF-S) observations of 23 square degrees at wide field depths in the first quick data release (Q1). Aims. With the goal of releasing complementary millimeter-wave data and encouraging legacy science, we performed dedicated observations of a 57-square-degree field overlapping the EDF-S. Methods. The observing time totaled 20 days, and we reached noise depths of 4.3, 3.8, and 13.2 $μ$K-arcmin at 95, 150, and 220 GHz, respectively. Results. In this work we present the temperature maps and two catalogs constructed from these data. The emissive source catalog contains 601 objects (334 inside EDF-S) with 54% synchrotron-dominated sources and 46% thermal dust emission-dominated sources. The 5$σ$ detection thresholds are 1.7, 2.0, and 6.5 mJy in the three bands. The cluster catalog contains 217 cluster candidates (121 inside EDF-S) with median mass $M_{500c}=2.12 \times 10^{14} M_{\odot}/h_{70}$ and median redshift $z$ = 0.70, corresponding to an order-of-magnitude improvement in cluster density over previous tSZ-selected catalogs in this region (3.81 clusters per square degree). Conclusions. The overlap between SPT and Euclid data will enable a range of multiwavelength studies of the aforementioned source populations. This work serves as the first step toward joint projects between SPT and Euclid and provides a rich dataset containing information on galaxies, clusters, and their environments.

Archipley, M. [Chicago U., Astron. Astrophys. Ctr.↗

Two-Stream Multi-Channel Convolutional Neural Network for Multi-Lane Traffic Speed Prediction Considering Traffic Volume Impact

Traffic speed prediction is a critically important component of intelligent transportation systems. Recently, with the rapid development of deep learning and transportation data science, a growing body of new traffic speed prediction models have been designed that achieved high accuracy and large-scale prediction. However, existing studies have two major limitations. First, they predict aggregated traffic speed rather than lane-level traffic speed; second, most studies ignore the impact of other traffic flow parameters in speed prediction. To address these issues, the authors propose a two-stream multi-channel convolutional neural network (TM-CNN) model for multi-lane traffic speed prediction considering traffic volume impact. In this model, the authors first introduce a new data conversion method that converts raw traffic speed data and volume data into spatial–temporal multi-channel matrices. Then the authors carefully design a two-stream deep neural network to effectively learn the features and correlations between individual lanes, in the spatial–temporal dimensions, and between speed and volume. Accordingly, a new loss function that considers the volume impact in speed prediction is developed. A case study using 1-year data validates the TM-CNN model and demonstrates its superiority. This paper contributes to two research areas: (1) traffic speed prediction, and (2) multi-lane traffic flow study.

97 MATHEMATICS AND COMPUTING↗

Bridging the Gap Between LLMs and LNS with Dynamic Data Format and Architecture Codesign

Deep neural networks (DNNs) have achieved tremendous success in the past few years. However, their training and inference demand exceptional computational and memory resources. Quantization has been shown as an effective approach to mitigate the cost, with the mainstream data types reduced from FP32 to FP16/BF16 and recently FP8 in the latest NVIDIA H100 GPUs. With increasingly aggressive quantization, however, the conventional floating-point formats suffer from limited precision in representing numbers around zero. Recently, NVIDIA demonstrated the potential of using a Logarithmic Number System (LNS) for the next generation of tensor cores. While LNS mitigates the hurdles in representing small numbers, in this work we observed a mismatch between LNS and the emerging Large Language Models (LLM), where LLM exhibits significant outliers when directly adopting the LNS format. In this paper, we present a data-format/architecture codesign to bright this gap. On the format side, we propose a dynamic LNS format to flexibly represent outliers at a higher precision, by exploiting asymmetry in the LNS representation and identifying outliers through a per-vector basis. On the architecture side, for demonstration, we realize the dynamic LNS format in a systolic array, which can handle the irregularity of the outliers at runtime. We implement our approach on an Alveo U280 FPGA as a prototype. Experimental results show that our design can effectively handle the outliers and resolve the mismatch between LNS and LLM, contributing to an accuracy improvement of 15.4% and 16% over the floating-point and the original LNS baselines, using four state-of-the-art LLM models. Our observation and design lay a solid foundation for the large-scale adoption of the LNS format in the next-generation deep learning hardware.

Haghi, Pouya↗

Condensed Phase Water Molecular Multipole Moments from Deep Neural Network Models Trained on Ab Initio Simulation Data

Ionic solvation phenomena in liquids involve intense interactions in the inner solvation shell. For interactions beyond the first shell, the ion–solvent interaction energies result from the sum of many smaller-magnitude contributions that can still include polarization effects. Deep neural network (DNN) methods have recently found wide application in developing efficient molecular models that maintain near-quantum accuracy. In this work, we extend the DeePMD-kit code to produce accurate molecular multipole moments in the bulk and near interfaces. The new method is validated by comparing the DNN moments with those generated by ab initio simulations. The moments are used to compute the electrostatic potential at the center of a molecular-sized hydrophobic cavity in water. The results show that the fields produced by the DNN models are in quantitative agreement with the AIMD-derived values. These efficient methods will open the door to more accurate solvation models for large solutes such as proteins.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

G0.253 + 0.016: A MOLECULAR CLOUD PROGENITOR OF AN ARCHES-LIKE CLUSTER

Young massive clusters (YMCs) with stellar masses of 10{sup 4}-10{sup 5} M{sub Sun} and core stellar densities of 10{sup 4}-10{sup 5} stars per cubic pc are thought to be the 'missing link' between open clusters and extreme extragalactic super star clusters and globular clusters. As such, studying the initial conditions of YMCs offers an opportunity to test cluster formation models across the full cluster mass range. G0.253 + 0.016 is an excellent candidate YMC progenitor. We make use of existing multi-wavelength data including recently available far-IR continuum (Herschel/Herschel Infrared Galactic Plane Survey) and mm spectral line (H{sub 2}O Southern Galactic Plane Survey and Millimetre Astronomy Legacy Team 90 GHz Survey) data and present new, deep, multiple-filter, near-IR (Very Large Telescope/NACO) observations to study G0.253 + 0.016. These data show that G0.253 + 0.016 is a high-mass (1.3 Multiplication-Sign 10{sup 5} M{sub Sun }), low-temperature (T{sub dust} {approx} 20 K), high-volume, and column density (n {approx} 8 Multiplication-Sign 10{sup 4} cm{sup -3}; N{sub H{sub 2}}{approx}4 Multiplication-Sign 10{sup 23} cm{sup -2}) molecular clump which is close to virial equilibrium (M{sub dust} {approx} M{sub virial}) so is likely to be gravitationally bound. It is almost devoid of star formation and, thus, has exactly the properties expected for the initial conditions of a clump that may form an Arches-like massive cluster. We compare the properties of G0.253 + 0.016 to typical Galactic cluster-forming molecular clumps and find it is extreme, and possibly unique in the Galaxy. This uniqueness makes detailed studies of G0.253 + 0.016 extremely important for testing massive cluster formation models.

79 ASTRONOMY AND ASTROPHYSICS↗

Chemical signature characterization with hyperspectral imagery: novel deep learning model architectures and physically-motivated data augmentation techniques

The high spectral resolution afforded by Hyperspectral Imaging (HSI) sensors is poised to bring unprecedented advancements to signature characterization applications. Thus far, much of the research in the machine learning field devoted to HSI applications has focused on a few specific tasks like land-use land-cover classification. In land classification tasks, spatial information is very important, and model architectures are often designed to leverage spatial contexts. However, it is unclear how well these spatially-tuned models will translate to tasks where spectral information is critical, like the detection and characterization of chemicals. In this work, we compare spectral models (inputs are 1D spectra) and spatial-spectral models (inputs are 3D cubes) in the context of predicting chemical concentration maps. We find that spatial-spectral models perform the best, though we find a wide range in performance across the different architectures tested. Additionally, we find that model performance is impacted by the availability of training data, particularly in scenarios where the training data doesn't fully capture the true variance of real-world conditions. We find that data augmentation can help mitigate sparse coverage of observed parameter space (e.g., seasonal or geographic variability in ground cover), and present augmentation strategies that are tailored to hyperspectral data.

• Artificial intelligence (AI) / machine learning ↗